Recently, with the rapid development and dissemination of network technologies, high-capacity content such as audio and video are being open to the public through a network. An example of the technology of delivering the content to the public is streaming delivery.
A client-server system is a general example of a network form that realizes large-scale streaming delivery. However, it inevitably requires expensive servers and building the network infrastructure therein, which results in a poor cost performance ratio.
For such a reason, as a network form that realizes large-scale streaming delivery without building up the expensive servers and the network infrastructure, there is known a technology using a peer-to-peer (hereinafter, referred to as P2P) system.
As an example of a network (hereinafter, referred to as a delivery network) that realizes streaming delivery through the application of the P2P system, a technique is known in which a terminal that has received data relays the received data to another terminal so as to perform sequential relay delivery to realize large scale delivery. This technique is advantageous in that the delivery load according to the increase in the number of terminals is less than that of the client-server system.
In a case where a new terminal is allowed to participate in this delivery network, it is necessary to search for a terminal (hereinafter, referred to as an optimal terminal for connection) that can continuously transmit data to the new terminal at a predetermined transmission rate to allow the new terminal to participate in the delivery network.
The known participation methods of a new terminal include a “hybrid-type P2P” system in which a server (hereinafter, referred to as a route management server) managing a delivery route determines a connection destination and instructs the new terminal to use the connection destination, a “pure-type P2P” system in which a connection destination is autonomously determined by searching between terminals not through a route management server, and the like.
In addition, as an intermediate method between the hybrid-type P2P and the pure-type P2P, there are known JXTA2 (P2P technology produced by Sun Microsystems) or a super node-type (or a super peer-type) P2P that is employed by skype (registered trademark) of an IP telephone service.
Furthermore, there are also known techniques of using a P2P system network which include a technique (for example, see Japanese Laid-open Patent Publication No. 2006-236344) of controlling access to a chat room and a technique (for example, see Japanese Laid-open Patent Publication No. 2005-109539) of searching for content information.
However, according to the above-described conventional technology, there are problems in that it is impossible to search for an optimal terminal with the amount of traffic of data, which is transmitted and received to determine a terminal to be connected, being appropriately suppressed, and it is difficult to build up a large-scale P2P network.
For example, in the hybrid-type P2P, as the terminal increases in number, the processing load of the route management server which to determine a connection destination and to manage the terminals increases. This will be described in detail with reference to FIG. 12. FIG. 12 is a diagram illustrating a problem with the hybrid-type P2P.
Here, a “new participating terminal” illustrated in FIG. 12 represents a terminal that newly participates in a delivery network. In addition, member terminals 1 to N represent terminals that transmit or receive content data such as audio or video through P2P communication. A “route management server” represents a server that manages terminal information of the member terminals 1 to N and selects a plurality of member terminals that are optimal for the connection of the new participating terminal and notifies of the selected member terminals.
In the hybrid-type P2P as illustrated in FIG. 12, when participation requests transmitted from new participating terminals are concentrated on the route management terminal and thus the traffic increases, the route selecting capability of the route management terminal becomes a bottle neck, and there is a problem in that it is difficult to realize large-scale streaming delivery.
On the other hand, according to the pure-type P2P, in a case where the number of terminals increases within a delivery network, in order to search for a terminal that is optimal for the connection, the amount of traffic of data transmitted for the search explosively increases. On the other hand, in a case where the amount of traffic for the search is limited, acquiring a terminal that is optimal for the connection is likely to be difficult.
This will be described in detail with reference to FIG. 13. FIG. 13 is a diagram illustrating a problem with the pure-type P2P. According to the pure-type P2P illustrated in FIG. 13, there is a problem in that the amount of traffic explosively increases in a case where a participation request transmitted from a new participating terminal is successively transmitted between the member terminals. On the other hand, in a case of limiting the number of times a new participating terminal transmits the participation request, a terminal that is optimal for the connection cannot be reliably found. Accordingly, there is a problem in that it is difficult to realize large-scale streaming delivery.
In addition, as an example of the intermediate method between the hybrid-type P2P and the pure-type P2P, there is a method in which a plurality of super nodes corresponding to route management terminals is provided, and terminals as search targets are determined based on a result of communications between the super nodes. After the terminals as the search targets are found, the super nodes end the search process.
Accordingly, even in a case where the intermediate method between the hybrid-type P2P and the pure-type P2P is used, each super node ends a search process at timing when the terminal as a search target is found, and therefore, the object of selecting several optimal terminals from among a plurality of candidates satisfying a condition is not achieved, and as a result, there is a problem in that it is difficult to realize large-scale streaming delivery.